Qwen3-8B-abliterated
8BQwen
from Qwen/Qwen3-8B
↓ 32.9K♥ 5826 repos
Qwen3-8B-abliterated-v2
Qwen3-8B-abliterated-v2 is a refusal-ablated variant of Qwen/Qwen3-8B, published on Hugging Face by huihui-ai. It is 8B parameters, single consumer GPU class and apache-2.0 licence. It is distributed across 17 repositories in Transformers, GGUF, MLX and GGUF (imatrix) formats, totalling 44.2K downloads.
Specification
- Model name
- Qwen3-8B-abliterated-v2
- Source repository
- huihui-ai/Huihui-Qwen3-8B-abliterated-v2
- Base model
- Qwen/Qwen3-8B
- Publisher
- huihui-ai
- Parameters
- 8B
- Hardware class
- 4B – 10B — single consumer GPU
- Licence
- apache-2.0
- Task
- Text generation
- Context window
- Not documented
- Ablation scope
- Not stated by the publisher
- Formats available
- Transformers, GGUF, MLX, GGUF (imatrix), FP8
- First indexed
- 18 Jun 2025
- Last updated
- 22 Aug 2026
Ablation technique
How Qwen3-8B-abliterated-v2 was modified, and what that implies.
Abliteration (directional ablation)
The single residual-stream direction that mediates refusal is identified from harmful/harmless prompt pairs and projected out of the model weights.
Reported scope: Not stated by the publisher. Publishers frequently omit this, so verify behaviour empirically rather than assuming full coverage.
From the publisher’s model card
“This is an uncensored version of Qwen/Qwen3-8B created with abliteration (see remove-refusals-with-transformers to know more about it). This is a crude, proof-of-concept implementation to remove refusals from an LLM model without using TransformerLens.”
Running it
Commands are templates — confirm the exact repository and quant file before use.
Ollama (publisher-documented)
ollama run huihui_ai/qwen3-abliterated:8b-v2llama.cpp / GGUF
hf download mradermacher/Huihui-Qwen3-8B-abliterated-v2-GGUF --local-dir ./qwen3-8b-abliterated-v2
llama-cli -m ./qwen3-8b-abliterated-v2/<file>.gguf -p "..."Transformers
from transformers import AutoModelForCausalLM, AutoTokenizer
repo = "huihui-ai/Huihui-Qwen3-8B-abliterated-v2"
tok = AutoTokenizer.from_pretrained(repo)
model = AutoModelForCausalLM.from_pretrained(
repo, torch_dtype="auto", device_map="auto"
)vLLM
vllm serve huihui-ai/Huihui-Qwen3-8B-abliterated-v2 --trust-remote-codeDownloads and variants (16)
Every published repository of Qwen3-8B-abliterated-v2, including quantised re-releases by other authors.
Security-team assessment
Derived from this model’s metadata and the publisher’s claims — not from benchmarks run by this site.
- Qwen3-8B-abliterated-v2 will attempt offensive-security prompts that a hosted commercial model declines, which is what makes it usable for red-team corpus generation and for measuring what an unaligned model of this class produces.
- Capability is inherited from Qwen/Qwen3-8B, not added by ablation — at 8B parameters, expect a small model’s reasoning and a high error rate on exploit detail.
- The publisher does not state which layers were ablated, so assume nothing about refusal consistency — probe it directly.
- Declared licence is apache-2.0, but the terms that bind you are Qwen/Qwen3-8B’s — a re-release cannot grant more than its parent.
Handling: run it isolated, put moderation in front of it if anyone outside your team can reach it, and verify the weights before loading — these are third-party artifacts. Full handling and licence guidance.
Metadata for Qwen3-8B-abliterated-v2 is collected from the public Hugging Face API and the publisher’s model card. This site does not host weights, is not affiliated with huihui-ai, and does not independently verify publisher claims. See responsible use.